So how can it be intuitive for a single line to represent an arbitrary range of quantities? Three apples exist in nature. There is evidence that the human brain can recognise the quantity. In fact, I believe dogs and parrots have been shown to have an internal concept of "3". But where does a concept like the number line occur in nature...?
To be fair though, they should all get basic concepts like the battery meter on a smartphone or the gas gauge on your car... or a glass being about a 1/3 full
Key concept: the glass is "one third full" -- IE fraction of "one", the whole line is "one". The battery, gas tank or glass is "1/3 full" -- this is not a numberline.
Similarly, human linguistics generally concerns itself only with mappings between symbolic concepts with no thought as to how those are truly internally represented nor how synthesis into external representation occurs.
No it most certainly does not. That's "semiotics". "Linguistics" is a much broader field.
No, it's more to do with the relative humidity/aridity of the local environment. There aren't many watery tarts in the Sahara, but a swimming pool at a Club 18-30 resort will be full of 'em.
While specialized texts may need to be rewritten every three years, I'm not sure that would apply to other works.
That's my point -- non-specialist texts don't need to be rewritten every three years, but at present they are. It seems to be specifically designed to stifle the trade in second-hand texts. This wasn't a problem when I studied in Edinburgh, because the lecturers set their own exercises, and were therefore happy to list both old and new edition section numbers in further reading, but I hear a lot of people complaining that under the US system, you need the correct edition of a textbook to follow a course.
It's not -- it just needs to be paid for. Peer review is effectively part of the creative process, so people aren't going to do it for free for a commercial publisher. And people will be particularly unlikely to do it for free if it's going to be rewritten every three years, resulting in the need for re-reviewing.
Of course, money corrupts, and there will be pressure on paid reviewers to make sure they don't sink a project. Free reviewers can be much more forthright.
I don't see the open textbook model replacing truly specialised texts, but for the fundaments of any fields of study, it shows great promise.
If anything, the lack of something is very fundamental to our existence. It's because we have zero meat we go hunt.
And yet you would not say that if I hadn't prompted you to.
Zero was a technology that was not properly understood, so people didn't use it. That's perfectly natural. There are plenty of examples of this. Don't drop it at the door of religion.
(And for the record, no, I don't believe in God/a god/gods/divine enlightenment/reincarnation/life after death. I just believe that most religion bashing is ignorant and unjustified.)
I think you (and your respondents) are missing the meaning of "peer review". We're not talking about Amazon reviews here ("The font's ugly, lolz") but about academic scrutiny. It's "a peer review" not "a review".
The main thing they forget to say on that wiki page is that the Romans counted their years as ordinal numbers (ie 1st, 2nd, 3rd), just like we do with days. There is no "zeroeth" except in certain notations of convenience within computer science. (Although to me x[0] will always be the first element of the array....)
Mathematics was held back quite a bit for quite a long time by religion. When institutionalized superstition abhors the concept of void (zero), you have a serious drawback.
(This is why 1BC is followed by 1AD, by the way.)
Oooh! Religion bashing! How unique and original.
The Greek philosophers spent a heck of a long time asking themselves whether something could be nothing. There is a fundamental difficulty, therefore, in naming something as "0". The first year of the Julian calendar existed. How can you call it nothing? Look at very young babies. Do we call them "zero-year olds"? No, we call them 4-week olds, three month olds etc, because we have to call them something. And how many brothers do I have? I have three. How many mothers? I only have one. How many dogs? I don't have any. You can't say "I have zero." You don't say "I have zero." Zero is not, and never has been, a genuine "number" to us psychologically. Of course it took a long time for people to accept it.
A reference text by definition doesn't "help" explain things; it describes them and leaves it up to the student to parse the content. My experience with my CS, math, and biology textbooks has been that they spend an absolute ton of time trying to explain the concepts in them as thoroughly as possible, using the most accessible analogies and descriptions they can. They're much easier to digest than a pure description would be—perhaps you just got the bad pick from the barrel if your books are just questions?
My most recent textbook experience was for an anatomy and physiology course I recently did. I felt it fell completely between the two stools -- it tried to be both a reference text, but attempted to teach, coursebook like.
The reason this didn't work is that if something is designed to be dipped in and out of, you're left making a lot of assumptions about existing knowledge. Most reference texts assume too much knowledge, making them thick and imprenetrable, but this one went out of it's way to explain various bits as pieces, leading it to be long and slow. Worse, there was often too much information, too much detail. Nothing was abstracted to a suitable level. For example, blood-flow through the heart was shown on an anatomically correct diagram -- but that meant that everything was too close together. In order to work out what was going on, I had to devise my own functional diagram....
Almost. Linguistics isn't just "absent", it's actively rejected by the establishment as being of zero utility to the learner. We're not allowed to stand on the shoulders of giants in the language class -- we're just supposed to stand at their feet and try to make out what sort of hat they're wearing....
And they run to Google when faced with horrible problems like "cook at 250 C" on a stove with F temperatures, because doing 1.8 x + 32 is beyond them.
Hey, I resent that remark. I can handle the arithmetic, but I just don't have any reason to remember the conversion, because it's Not That Important. I spend far more time converting between different bases than different measurement systems.
There's a thousand Algebra books because there are a thousand times that number of teachers and all have their own preferences. Throughout school and university, I never viewed a book as anything more than a recommendation and I was forced to buy precisely ONE book (and that because the teacher set exercises by page number, which is nothing to do with the book itself - but it does make you wonder who got the back-hander).
You were forced to buy one book? That means you were given the responsibility for buying books -- lucky you. I think you'll find that a great many teachers go their entire careers without ever getting a say on the books they use in class.
I think the biggest advantage of this project will turn out to be not the price, as everyone would expect, but the academic rigour in the review process. Given the horror stories floating around about the poor standards of review boards for high school textbooks, I can only imagine the delight of teachers seeing a textbook with a university's seal of approval on it.
Downmodded. How disingenuous of a site with so many programmers who know firsthand of the shit that comes out of India.
If I had any mod points I would mod you up as "insightful". Because I wasn't aware that "Sunnyvale, Calif." was in India. I thought it was in the US. Thanks for clearing that up.
Aha, so the they're taking one monolithic source program, identifying what can be done in software, what standard components are required in order for the code to function properly, and what non-standard stuff has to be custom designed... and then designing it. ++nontrivial. Very cool tech.
I'm not entirely clear on how it works though. If I give them this:
#include <stdio.h>
int main() {
printf("Hello world!\n");
}
they will convert it into a custom integrated circuit chip with Hello World! silkscreened on the top of it or does the chip actually display "Hello World!" on whatever it is connected to?
It won't do anything. The above piece of code will not translate to synthesizable logic. The printf() statement is not synthesizable. For the tool to output something meaningful, the input has to be meaningful (to the tool).
But we're not just talking about FPGA imaging here -- we're talking system-on-a-chip with processor, RAM, Flash etc. So their system will generate a SoC from stock components with a processor, minimal Flash RAM and a very basic GPU. They won't need to generate any custom logic, but they'll still be generating a SoC to spec, one that will connect to some video display and show a single message.
The point is that one person writes the summary and a huge number of people read it. Not all of them know everything.
Exactly. Or to put it in a way any geek should understand...
Your "source code" (= the summary) refers to an external source (= the definition of the abbreviation SoC). You could use the preprocessor to fetch that and insert it into the code as a string literal before running compilation (slashdot submission). As you didn't, this meant that the object code generated (= the slashdot summary) invoked a remote file access operation where the host operating system (= the slashdot reader's brain) didn't currently have the string literal "system on a chip" in local RAM.
Substituting the literal at compile time would have removed the need for thousands of file reads at the cost of 13 * sizeof (char) bytes.
Any mistake in a SoC is expensive, especially if you go directly from design to wafer without extensive testing. Most of the time it's 1 week of actual writing a description in VHDL or Verilog and then spending a few weeks/months verifying the design and removing any bug.
See, this is what's confusing me. Isn't a system-on-a-chip just CPU+GPU+soundset+RAM+flash on one chip? Is there any real hardware to implement? The whole summary makes it sound like they've implemented... a C compiler.....
While it is true that you can engineer essays to be 'bad' and still score 'good' - the question is - are there natural essays that score good but are actually bad; and good essays that score bad but are actually good.
Don't misinterpret the results. Just because it wasn't a natural essay doesn't make it a bad example. The prof has shown what the system responds well/badly to, irrespective of mode of writing.
This whole thing reminds me of a summer school I took in Spanish -- the teacher did a prep session where we were rewarded for using a set of grammar points and connectors that the examiners would be looking for. We had a "good presentation" competition and my team won. Our presentation was practically content free, with sentences equivalent to: however, not withstanding the aformentioned problems, and bearing in
mind what has already been said, it can be seen that a solution would be difficult to implement. The other team called us cheats, the teacher said we followed the task perfectly.
Now I teach English as a foreign language for a living, and that experience has had a profound effect on me. My goal in exam preparation classes is to get my students to the point where they can write a "Smurf essay" -- something that matches the structure expected in the exam, but with absolutely zero content: This paper will look a the situation regarding the gradual smurphication of the smurph field. It will take into account three factors: the smurphiness of modern smurphs, the smurphicity of smurphism and the smurphology of smurphdom....
Mr. Perelman spent a month of effort carefully crafting an essay that said nothing, eloquently. If our students can do that, more power to them.
But if you read TFA to the end, you'd see this quote:
"Two former students who are computer science majors told him that they could design an Android app to generate essays that would receive 6’s from e-Rater."
...which kind of defeats the purpose of the exercise. Why would I spend a day trying to craft independent thought if I could get a guaranteed pass for a $0.99 download?
The marker bot doesn't reward "good writing", it rewards the employ of a few very superficial metrics. Which is like the language exams I've done.
I don't see why this would be different from current auditing practices. If an external examiner finds that your students have been incorrectly marked, it's either an automatic scaling of grades for everyone, or back to the red pen and regrade everything.
Another scary thought is that people are so dependent on software and computers they would not know what to do if for whatever reason there was an extended power outage or worse.
...which was precisely the problem at Fukushima. The emergency cooling system was electric, and the tsunami knocked out the emergency backup generator that powered it. There was a manual override, but everyone had forgotten that they needed to go and crank a few handles because as far as they were concerned, the system was "automatic". Untold millions of pound/euros/dollars/yen worth of material damage, hundreds of people put in direct physical danger from radiation, extensive contamination of the environs. Because it's "automatic".
Slashdot Mirror
Q: How many lines are there in a number line?
A: One.
So how can it be intuitive for a single line to represent an arbitrary range of quantities? Three apples exist in nature. There is evidence that the human brain can recognise the quantity. In fact, I believe dogs and parrots have been shown to have an internal concept of "3". But where does a concept like the number line occur in nature...?
To be fair though, they should all get basic concepts like the battery meter on a smartphone or the gas gauge on your car... or a glass being about a 1/3 full
Key concept: the glass is "one third full" -- IE fraction of "one", the whole line is "one". The battery, gas tank or glass is "1/3 full" -- this is not a numberline.
Similarly, human linguistics generally concerns itself only with mappings between symbolic concepts with no thought as to how those are truly internally represented nor how synthesis into external representation occurs.
No it most certainly does not. That's "semiotics". "Linguistics" is a much broader field.
Does "mikael" look like a native English-speaker's handle to you?
Right. There is no intrinsic link between intelligence and knowledge of English.
No, it's more to do with the relative humidity/aridity of the local environment. There aren't many watery tarts in the Sahara, but a swimming pool at a Club 18-30 resort will be full of 'em.
While specialized texts may need to be rewritten every three years, I'm not sure that would apply to other works.
That's my point -- non-specialist texts don't need to be rewritten every three years, but at present they are. It seems to be specifically designed to stifle the trade in second-hand texts. This wasn't a problem when I studied in Edinburgh, because the lecturers set their own exercises, and were therefore happy to list both old and new edition section numbers in further reading, but I hear a lot of people complaining that under the US system, you need the correct edition of a textbook to follow a course.
It's not -- it just needs to be paid for. Peer review is effectively part of the creative process, so people aren't going to do it for free for a commercial publisher. And people will be particularly unlikely to do it for free if it's going to be rewritten every three years, resulting in the need for re-reviewing.
Of course, money corrupts, and there will be pressure on paid reviewers to make sure they don't sink a project. Free reviewers can be much more forthright.
I don't see the open textbook model replacing truly specialised texts, but for the fundaments of any fields of study, it shows great promise.
If anything, the lack of something is very fundamental to our existence. It's because we have zero meat we go hunt.
And yet you would not say that if I hadn't prompted you to.
Zero was a technology that was not properly understood, so people didn't use it. That's perfectly natural. There are plenty of examples of this. Don't drop it at the door of religion.
(And for the record, no, I don't believe in God/a god/gods/divine enlightenment/reincarnation/life after death. I just believe that most religion bashing is ignorant and unjustified.)
I think you (and your respondents) are missing the meaning of "peer review". We're not talking about Amazon reviews here ("The font's ugly, lolz") but about academic scrutiny. It's "a peer review" not "a review".
The main thing they forget to say on that wiki page is that the Romans counted their years as ordinal numbers (ie 1st, 2nd, 3rd), just like we do with days. There is no "zeroeth" except in certain notations of convenience within computer science. (Although to me x[0] will always be the first element of the array....)
Mathematics was held back quite a bit for quite a long time by religion. When institutionalized superstition abhors the concept of void (zero), you have a serious drawback. (This is why 1BC is followed by 1AD, by the way.)
Oooh! Religion bashing! How unique and original.
The Greek philosophers spent a heck of a long time asking themselves whether something could be nothing. There is a fundamental difficulty, therefore, in naming something as "0". The first year of the Julian calendar existed. How can you call it nothing? Look at very young babies. Do we call them "zero-year olds"? No, we call them 4-week olds, three month olds etc, because we have to call them something. And how many brothers do I have? I have three. How many mothers? I only have one. How many dogs? I don't have any. You can't say "I have zero." You don't say "I have zero." Zero is not, and never has been, a genuine "number" to us psychologically. Of course it took a long time for people to accept it.
A reference text by definition doesn't "help" explain things; it describes them and leaves it up to the student to parse the content. My experience with my CS, math, and biology textbooks has been that they spend an absolute ton of time trying to explain the concepts in them as thoroughly as possible, using the most accessible analogies and descriptions they can. They're much easier to digest than a pure description would be—perhaps you just got the bad pick from the barrel if your books are just questions?
My most recent textbook experience was for an anatomy and physiology course I recently did. I felt it fell completely between the two stools -- it tried to be both a reference text, but attempted to teach, coursebook like.
The reason this didn't work is that if something is designed to be dipped in and out of, you're left making a lot of assumptions about existing knowledge. Most reference texts assume too much knowledge, making them thick and imprenetrable, but this one went out of it's way to explain various bits as pieces, leading it to be long and slow. Worse, there was often too much information, too much detail. Nothing was abstracted to a suitable level. For example, blood-flow through the heart was shown on an anatomically correct diagram -- but that meant that everything was too close together. In order to work out what was going on, I had to devise my own functional diagram....
Almost. Linguistics isn't just "absent", it's actively rejected by the establishment as being of zero utility to the learner. We're not allowed to stand on the shoulders of giants in the language class -- we're just supposed to stand at their feet and try to make out what sort of hat they're wearing....
And they run to Google when faced with horrible problems like "cook at 250 C" on a stove with F temperatures, because doing 1.8 x + 32 is beyond them.
Hey, I resent that remark. I can handle the arithmetic, but I just don't have any reason to remember the conversion, because it's Not That Important. I spend far more time converting between different bases than different measurement systems.
There's a thousand Algebra books because there are a thousand times that number of teachers and all have their own preferences. Throughout school and university, I never viewed a book as anything more than a recommendation and I was forced to buy precisely ONE book (and that because the teacher set exercises by page number, which is nothing to do with the book itself - but it does make you wonder who got the back-hander).
You were forced to buy one book? That means you were given the responsibility for buying books -- lucky you. I think you'll find that a great many teachers go their entire careers without ever getting a say on the books they use in class.
I think the biggest advantage of this project will turn out to be not the price, as everyone would expect, but the academic rigour in the review process. Given the horror stories floating around about the poor standards of review boards for high school textbooks, I can only imagine the delight of teachers seeing a textbook with a university's seal of approval on it.
Downmodded. How disingenuous of a site with so many programmers who know firsthand of the shit that comes out of India.
If I had any mod points I would mod you up as "insightful". Because I wasn't aware that "Sunnyvale, Calif." was in India. I thought it was in the US. Thanks for clearing that up.
Aha, so the they're taking one monolithic source program, identifying what can be done in software, what standard components are required in order for the code to function properly, and what non-standard stuff has to be custom designed... and then designing it. ++nontrivial. Very cool tech.
I'm not entirely clear on how it works though. If I give them this:
#include <stdio.h> int main() { printf("Hello world!\n"); }
they will convert it into a custom integrated circuit chip with Hello World! silkscreened on the top of it or does the chip actually display "Hello World!" on whatever it is connected to?
It won't do anything. The above piece of code will not translate to synthesizable logic. The printf() statement is not synthesizable. For the tool to output something meaningful, the input has to be meaningful (to the tool).
But we're not just talking about FPGA imaging here -- we're talking system-on-a-chip with processor, RAM, Flash etc. So their system will generate a SoC from stock components with a processor, minimal Flash RAM and a very basic GPU. They won't need to generate any custom logic, but they'll still be generating a SoC to spec, one that will connect to some video display and show a single message.
What does "/." stand for?
It's a redundant alias for the system root directory. Assuming you're using a proper computer and not one of these modern toys.
\me ducks
The point is that one person writes the summary and a huge number of people read it. Not all of them know everything.
Exactly. Or to put it in a way any geek should understand...
Your "source code" (= the summary) refers to an external source (= the definition of the abbreviation SoC). You could use the preprocessor to fetch that and insert it into the code as a string literal before running compilation (slashdot submission). As you didn't, this meant that the object code generated (= the slashdot summary) invoked a remote file access operation where the host operating system (= the slashdot reader's brain) didn't currently have the string literal "system on a chip" in local RAM.
Substituting the literal at compile time would have removed the need for thousands of file reads at the cost of 13 * sizeof (char) bytes.
Any mistake in a SoC is expensive, especially if you go directly from design to wafer without extensive testing. Most of the time it's 1 week of actual writing a description in VHDL or Verilog and then spending a few weeks/months verifying the design and removing any bug.
See, this is what's confusing me. Isn't a system-on-a-chip just CPU+GPU+soundset+RAM+flash on one chip? Is there any real hardware to implement? The whole summary makes it sound like they've implemented... a C compiler.....
While it is true that you can engineer essays to be 'bad' and still score 'good' - the question is - are there natural essays that score good but are actually bad; and good essays that score bad but are actually good.
Don't misinterpret the results. Just because it wasn't a natural essay doesn't make it a bad example. The prof has shown what the system responds well/badly to, irrespective of mode of writing.
This whole thing reminds me of a summer school I took in Spanish -- the teacher did a prep session where we were rewarded for using a set of grammar points and connectors that the examiners would be looking for. We had a "good presentation" competition and my team won. Our presentation was practically content free, with sentences equivalent to: however, not withstanding the aformentioned problems, and bearing in mind what has already been said, it can be seen that a solution would be difficult to implement. The other team called us cheats, the teacher said we followed the task perfectly.
Now I teach English as a foreign language for a living, and that experience has had a profound effect on me. My goal in exam preparation classes is to get my students to the point where they can write a "Smurf essay" -- something that matches the structure expected in the exam, but with absolutely zero content: This paper will look a the situation regarding the gradual smurphication of the smurph field. It will take into account three factors: the smurphiness of modern smurphs, the smurphicity of smurphism and the smurphology of smurphdom....
Mr. Perelman spent a month of effort carefully crafting an essay that said nothing, eloquently. If our students can do that, more power to them.
But if you read TFA to the end, you'd see this quote:
"Two former students who are computer science majors told him that they could design an Android app to generate essays that would receive 6’s from e-Rater."
...which kind of defeats the purpose of the exercise. Why would I spend a day trying to craft independent thought if I could get a guaranteed pass for a $0.99 download?
The marker bot doesn't reward "good writing", it rewards the employ of a few very superficial metrics. Which is like the language exams I've done.
I don't see why this would be different from current auditing practices. If an external examiner finds that your students have been incorrectly marked, it's either an automatic scaling of grades for everyone, or back to the red pen and regrade everything.
Another scary thought is that people are so dependent on software and computers they would not know what to do if for whatever reason there was an extended power outage or worse.
...which was precisely the problem at Fukushima. The emergency cooling system was electric, and the tsunami knocked out the emergency backup generator that powered it. There was a manual override, but everyone had forgotten that they needed to go and crank a few handles because as far as they were concerned, the system was "automatic". Untold millions of pound/euros/dollars/yen worth of material damage, hundreds of people put in direct physical danger from radiation, extensive contamination of the environs. Because it's "automatic".